The targets of most therapeutic antibodies are found on healthy cells which limits the development of truly tumor-specific therapies and results in a narrow therapeutic index. Using state-of-the-art proteomic and gene profiling methods, our collaborators have identified unique sets of tumor-specific cell surface antigens that are not usually co-expressed on non-malignant cells. These findings support the development of a novel binary-weapon therapy wherein, unlike conventional bispecific antibodies, our antibody fragments dimerize at the tumor cell surface ONLY IF both targets are present at the surface of the same cell (Christopherson, 2006; Ellmark, 2008; Barber, 2009; Kohnke, 2009; Kaufman, 2010). To this end we will develop Demibodies hetero-dimerization-activated 6 antibodies. Each demibody consists of 3 molecular components: (1) a single-chain variable fragment (scFv) to target a specific tumor cell-associated surface antigen, (2) half of a leucine zipper (E3 or K3) to dimerize the fragments, and (3) half of a binary toxin to mediate tumor cell-specific cytotoxicity. Members of the prototypic demibody pair (A, B) do not dimerize when free in solution. Each component of a demibody pair binds to a different target surface antigen that is uniquely expressed on the tumor cell surface. Dimerization is facilitated by a) the high local concentrations of the demibodies at the cell surface, b) the constraining of the movement of each demibody to two dimensions at the plasma membrane, and c) the stabilization afforded by association of the complementary halves of the leucine zipper. Upon formation of the demibody heterodimer, genetically linked toxin domains cooperate to induce cellular cytotoxicity. For proof-of-principle, a demibody pair that did not interact in solution showed complementary cell surface binding, as demonstrated by Forster resonance energy transfer (FRET). In Phase I, we will evaluate cytotoxicity of CD5:CD19 demibody pair that carries the potently synergistic immunotoxins of the innate immune system, granzyme B and perforin. We will test this novel therapeutic approach as a potential immunotherapy for chronic lymphocytic leukemia (CLL). PUBLIC HEALTH RELEVANCE: Chronic lymphocytic leukemia (CLL) is the most common leukemia in the US. Current therapies for CLL have limited efficacy and significant side effects. We have designed activatable antibodies that recognize combinations of antigens found uniquely on cancer cells. Following dimerization on the cell surface, these demibodies activate cytotoxins and result in specific killing of cancer cells. Development of demibody technology will enable a new class of cancer therapeutics with increased efficacy and fewer side effects.